Abstract
Insects not only spread diseases to humans and affect the quality of life or health, but also affect food factories’ operation. Therefore, the identification of various types of insects becomes extremely important in epidemic prevention.
The ultimate purpose of this study is to establish an insect species identification system, which can be divided into two main parts, object detection and insect species identification. To begin with, YOLO-v4 deep learning model is used to detect insects in the entire image which is produced by shooting the sticky traps. Then, the regional image of a single insect is extracted through the bounding box generated by YOLO-v4. Finally, the GoogLeNet Inception-v4 deep learning model is used to identify a species of insect in the regional image. Also, in order to improve the accuracy, image processing will be carried out before inputting the image into the model. Additionally, a combined model will be used to solve the confusion caused by the species identification model for some species look alike.
In the final results of this study, the recall value of the object detection model is 96%, and the accuracy rate of the species identification model is 87.1%.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bochkovskiy, A., Wang, C.-Y., Liao, H.-Y.M.: YOLOv4: Optimal Speed and Accuracy of Object Detection
Neubeck, A., Van Gool, L.: Efficient non-maximum suppression. In: 18th International Conference on Pattern Recognition (ICPR 2006), pp. 850–855 (2006). https://doi.org/10.1109/ICPR.2006.479
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J.: Rethinking the Inception Architecture for Computer Vision
Lin, M., Chen, Q., Yan, S.: Network in network
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: A Simple Way to Prevent Neural Networks from Overfitting
Rahman, S., Rahman, M.M., Abdullah-Al-Wadud, M., Al-Quaderi, G.D., Shoyaib, M.: An adaptive gamma correction for image enhancement (2016)
Szegedy, C., Ioffe, S., Vanhoucke, V., Alemi, A.A.: Inception-v4, inception-resnet and the impact of residual connections on learning. In: Thirty-First AAAI Conference on Artificial Intelligence, February 2017
WHO: Vector borne diseases. (https://www.who.int/zh/news-room/act-sheets/detail/vectorborne-diseases)
Wikipedia: Clahe. (https://en.wikipedia.org/wiki/Adaptive_histogram_equalization#CLAHE)
Wikipedia: YUV. (https://zh.wikipedia.org/wiki/YUV)
Zhang, X.-Y.: Convolutional Neural Network Based Mosquito Species and Gender Identification System, National Chung Hsing University, Department of Management Information System (2020)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Chan, YK., Lee, WC., Chen, WX., Chen, YC., Tu, WC., Yeh, ZH. (2022). Insect Species Identification System Based on Deep Learning. In: Lee, R. (eds) Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing. SNPD 2021. Studies in Computational Intelligence, vol 1012. Springer, Cham. https://doi.org/10.1007/978-3-030-92317-4_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-92317-4_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-92316-7
Online ISBN: 978-3-030-92317-4
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)